Visualizing Disaster: Remote Sensing, Boundaries, and Invisibility

In the wake of Japan’s 8.9 earthquake on 11th March, 2011, global media attention has shifted dramatically from the ongoing political unrest in the Middle East to breaking news of both explosions at Fukushima Daiichi and failures at Fukushima Dain—two of Japan’s nuclear facilities—and the increasing threat of radiation contamination. An international community that is concerned and anxious for information has been bombarded with a bevy of images. This in itself is nothing unusual: in the wake of any natural or anthropogenic disaster, imagery of victims and devastation are played and replayed continuously. Increasingly, with both the ubiquity of high-resolution satellite imagery coverage, private corporations, such as Digital Globe and GeoEye are providing free high-resolution imagery to aid in disaster relief. The recent usage of satellite imagery in both the Haitian and the Gulf oil spill disasters highlights the applicability of this technology in natural and anthropogenic disasters. In Haiti, for example, the impact of the earthquake on rural areas rendered inaccessible by destroyed roads and bridges was made legible through satellite imagery. For critical geographers, the usage of remotely sensed imagery in the subsequent British Petroleum oil spill in the Gulf of Mexico was particularly fascinating as it was through coarse scale satellite images and fine scale underwater videos that the disaster was re-produced for concerned citizens, politicians, disaster management responders, and corporate executives. Writing about the BP oil spill online at Placing Culture, at that time, I asked the question on many geographer’s minds: “whether at a broad or fine scale, what do these media forms, and their usages, tell us about interactions between humans and the environment in this contemporary moment. More specifically, how do these technologies, and their usages, inform our understanding of the "production" of environmental disasters, their (re)mediation, and the assignment of blame?” Confronted with a very different series of concatenated disasters—an earthquake, tsunami, and nuclear plant explosion—I believe that critical geographers should ask different questions, ones about how the boundaries between visibility, invisibility, and legibility are mediated remotely through satellite imagery.

Over the last two decades, post-modern geopolitical debates have raged concerning the ostensible dissolution of territories, boundaries, and sovereignty in a globalized world where the internet and other communication media have caused a time-space compression. Following these debates, our understanding of boundaries as fixed delineators underwent radical changes: no longer the strict dividers of territory, boundaries are now seen not only as permeable, but, rather as social processes. Coterminous with these debates have been others in critical cartography surrounding the objectivity of remotely sensed imagery, and relatedly, whether geographic information systems traffic in data or more subjectively mediated information. In the context of these debates over boundaries, permeability, and objectivity, I believe it is fascinating to compare both media coverage, and conversely the (re)production, of natural disasters. Ironically, the intersecting point for analyses of visibility, invisibility, and legibility in the context of these disasters are the boundaries of radiation itself.

As a very brief introduction, remote sensing is the recording of data about an object from a sensor that is not in direct contact with the object. Similar in a way to photographic film, remotely sensed images are produced through the capturing of electromagnetic radiation. Geographers use remotely sensed imagery for a wide variety of purposes, including monitoring temporal changes in land cover, and modeling sea-level change. Critical geographers have also utilized and explored the usage of remotely sensed imagery, such as in the production and contestation of particular land change narratives. Other critical scholarship has explored the intersection of scale and boundaries in remotely quantifying spatial phenomena. Comparing the BP oil spill in the Gulf with the radiation leaks in Japan is particularly instructive, as it points to the centrality of borders as mediating the visibility, and conversely invisibility of certain types of disasters; additionally, such a comparison of borders also highlights the temporal variable through which certain visible disasters become invisible, and those that are invisible become visible. Take the BP oil spill, for example, for months satellite imagery was extensively utilized to help visualize and anticipate the spreading boundaries of the oil slick. With the spectral response of oil easily distinguishable from water, both scientists, and importantly the lay public were able to delineate and thus visualize the implications of the spill. In other words, the easily distinguishable natural boundaries of the spill enabled the visualization and production of the spill as a disaster. Ironically, given that electromagnetic radiation is the physical force that enables the production of remotely sensed imagery, radiation emanating from an explosion at nuclear reactor is a different story. Spectral sensors, at present, are not equipped to determine the spreading extent of this type of radiation. While thermal radiation, such as from a meltdown, for example, can be identified, radioactive isotopes that are air or waterborne are invisible. Or at least their current manifestations are invisible. What is particularly intriguing to me is how the variable of time becomes a factor in the scientific analysis, media portrayal, and the public understanding of visible and invisible disasters. Additionally, in the context of ongoing debates about the social production of borders, it is fascinating how boundaries both coalesce and disperse over time. It has been a quarter century since the nuclear meltdown at Chernobyl; yet, the invisible and toxic cloud of radiation that was carried in the air and through the water has over time left visible spatial artifacts. Using multi-temporal Landsat imagery for example, scientists are capable of tracking forest regeneration following the abandonment of agricultural land.




In other words, the boundaries of what was a largely invisible disaster have become visible. By contrast, over even short periods of time, the incorporation and dispersal of oil within the Gulf stream has rendered the highly visible BP oil spill largely invisible—if not to the most accurate spectral sensors, than at least to the public eye. What these ruminations and imagery point to is how certain geographic phenomena—both anthropogenic and natural—are made visible or are inherently invisible, if only temporally, before other natural processes transform their visibility. For critical geographers, the differential production of disasters as (in)discernible and (un)bordered phenomenon, and how remotely sensed imagery mediate the related public and institutional responses are increasingly fruitful grounds for continued attention.

Social Media, Resistance, and PPGIS

For those with an interest in the intersections of evolving forms of social media, geospatial technologies, place, and culture the recent wave of protests sweeping the Middle East is of particular interest.

As I wrote in a recent paper (Meek 2011), "the development of Internet technology in general, and new social media in particular, are tracking and facilitating the spatial evolution of social interactions; as geographic constraints on communication increasingly dissolve, new forms of collective action are arising that transcend traditional conceptions of cultural and spatial fixedness (Jordan 1999; Rheingold 2003; Pickerill 2004, 2007)."


Later on in that paper I noted "New forms of social media are increasingly becoming integrated into evolving forms of political protest (Adams 1996; Myers 1994; Froehling 1997; Pickerill 2004; Mamadouh 2004; Juris 2005). Flash mobs (also known as smart mobs) are exemplary of the political usage of new social media along the continuum of place and cyberplace (Rheingold 2003; Gore 2010; Molnár forthcoming). In essence, flash mobs are relatively impromptu social gatherings in urban spaces that are organized using social media. This movement, which developed from the “Reclaim the Streets” movement in the late 1990s, employs mass text messaging and email lists to organize pseudo-spontaneous raves, political demonstrations, and performance art experiences. In studying the links between flash mobs and mobilization, Molnár (forthcoming) argues that “the emergence and immense popularity of flash mobs forcefully demonstrate that the virtual and the physical are not parallel realms but continuously intersecting social realities” (5). This continuity between online-offline collective action stands in stark contrast to previous dichotomies between physical space and cyberspace, and buttresses arguments for seeing place as a continuum (Wellman 2001; Wilson 2006; Broadfoot et al. 2010). Examining the Philippine’s Power People II movement, for example, which mobilized through text messages to overthrow the Estrada regime in 2001 (Lande 2001; Molnár forthcoming), one finds it difficult to disentangle where resistance occurred. Rather, than specifying these actions as either occurring in space or cyberspace, seeing them located along a continuum of place is more instructive, pointing us toward their embodied interconnections. Kluitenberg (2006), for example, focuses on these reticulations, arguing that flashmobs “demonstrate that we are living in a space in which the public is reconfigured by a multitude of media and communication networks interwoven into the social and political functions of space to form a ‘hybrid space’ (8).” Within this hybrid space, Molnár found that in contemporary flash mobs, YouTube was the primary point of entry for participants, as the spreading of awareness through viral media is integral to these forms of collective action gaining a critical mass. This brief illustration of flash mobs is instructive of the importance of examining the role of evolving social media in attempts to situate collective action."

Given these consideration, it should be no surprise with the recent unrest in Tunisia, Egypt, Bahrain, Libya etc blogs, wikis, wikilieaks, Facebook, Youtube, and other Web 2.0 application are increasingly become commonplace within the landscape of resistance.

Similarly, as I note in the above passages, given that social connections are em-placed, even within cyberplace, we should expect geospatial web applications as the obvious next social stone in the proverbial slingshot. A recent article on Wired (linked here) describes the usage of a Google Earth map populated by Twitter account of resistance and conflict in Libya. Interestingly, the map doesn't dynamically re-present the occurrence and locations of these events online--as that would be a security concern for the individuals on the ground.


View Mapping Violence Against Pro-Democracy Protests in Libya in a larger map

Perusing this map is not only sobering, but incredibly stimulating in re-thinking the intersection of social media, geospatial technologies, and resistance. What it points to at the most general is the importance and power of what's termed "user-generated content". User-generated content includes anything from YouTube videos to Google Earth maps such as the one above They highlight what scholars in Media 2.0 studies have termed the shift to the "prosumer". More specifically, it directs our attention to the myriad potential intersections of social media, geospatial technology, place, culture, and resistance. With the increasing ubiquity of wireless internet, smart phones, free social media applications like Facebook, YouTube, Twitter, and geospatial ones such as Google Earth it is anyone's guess where the next protests will spring up...if the above ruminations are any indication, it will most surely be emplaced somewhere along the continuum of place and cyberplace.


During the previous generation of media studies, cyberspace was largely defined by the dissolution of traditional spatial boundaries (Mitra and Watts 2002). These descriptions helped cement a dichotomy between actions taken in space versus those in cyberspace (Castells 1996; Miller and Slater 2000). Since 2004, with the origination of Web 2.0, and Media 2.0 studies, this distinction has begun to dissolve (Kluitenberg 2006; Guantlett 2009; Valtysson 2010). As Manovich (2009) argues, we have moved from media to social media. This shift is exemplified by the explosion of user-generated content, such as blogs and wikis, and the transition from media consumption to prosumption (Klinenberg 2005; Humphreys 2008; Grinnel 2009, Napoli 2010; Ritzer and Jurgenson 2010; Paasonen 2010; Valtysson 2010).

(The paper I mentioned is Meek, D. 2011. “YouTube and Social Movements: A Phenomenological Analysis of Participation, Events, and Cyberplace”. In revision- Antipode."

Introducing Techniques Tutorials

Some of the impediments to becoming proficient with GIS and remote sensing for many people is the complexity of the software systems, and time necessary to follow the learning curve.

Through adding a section to this blog with GIS and remote sensing screencasts, I endeavor to help people become more familiar with the techniques of GIS, remote sensing, and spatial analysis. Below is the first screen cast. More will be posted on the Techniques pages on the right. All videos can also be found by searching for Placingculture on Youtube. Enjoy, and please let me know of any comments, criticisms, or suggestions for subsequent screencasts.

Localisation, Participation and Communication PGIS

This great video was posted by Giacomo Rambaldi on the PPGIS blog.

Here is the description from the blog:

The 25-min educational video documentary (see link below) introduces the practice of participatory spatial information management and communication (PGIS) in the development context. It has been designed to introduce development practitioners (technology intermediaries) to the practice of demand-driven PGIS.

In this video, PGIS practice is presented as a continuum starting from community mobilisation to project planning and design, choice of mapping methods and technologies, visualisation of different technologies in diverse ethno-cultural and agro-ecological environments, and finally putting the maps to work in the domains of identity building, self-determination, spatial planning and advocacy.

Ethics and sound attitudes and behaviours are emphasized as cross-cutting imperatives.

Localisation, Participation and Communication: an Introduction to Good PGIS Practice from CTA on Vimeo.

Evolving Data Server Solutions

Many are interested in the intersection of geospatial technologies, place, and culture because of the power that maps have for telling stories about the relationships between people and the places they inhabit and transform. That is certainly one reason I´m continually fascinated by this subject. However, if once produced a map simply sits in a dusty filing cabinet than what good is it?

Geographic information systems (GIS) as digital repositories of spatial data (including both quantitative and qualitative data forms, vis. a vis sources such as multimedia and embedded text). With the advent and now ubiquitous place of GIS it has become very easy to transmit maps and cartographic data electronically.

The caveat, until relatively recently, has been how and to whom the data are transmitted. Emailing a map as a .jpg or set of cartographic data as a zipped attachement to a colleague is one thing. Making those data available for public usage is completely different.

High-end (in terms of both financial expenditure and tehnological prowess) spatial data servers have been online for sometime now. These include product by the big names in databases Oracle for example. ESI´s ArcGIS server is another common player. However, these are mostly spatial database solutions for those with big budgets (again in terms of finances and labor-which are maybe the same...).

What options are there for those of us interested in more open-source data serving solutions?

One option that I'd like to highlight is called Data Basin. It is a great site that allows users to "explore and download a vast library of datasets, connect to external data sources, upload and publish your own datasets, connect to experts, create working groups, and produce customized maps that can be easily shared". And best of all....it's free! Within Data Basin, users can search and download datasets, as well as share their own. The website also provides a space for user groups where spatial data can be easily shared. Very cool indeed.

Mapping the Deep Water Horizon

The basic details are well known: on April 20th, an explosion on the drilling rig "Deepwater Horizon" killed eleven crewmen; the fire was could not be extinguished and on April 22nd the rig sank, leaving the well gushing, resulting in the largest off-shore oil spill in United States history.

Since late April, millions across the globe have become increasingly concerned with the impending disaster growing in the Gulf of Mexico. One of the factors behind this growing concern have been the proliferation and extensive usage of remote sensing technologies. From satellite sensors to underwater cameras, concerned citizens, politicians, disaster management responders, and corporate executives have anxiously followed the online images and video feeds.



Aside from pretty (or grainy) pictures and videos, at a broad scale, what do these media forms, and their usages, tell us about interactions between humans and the environment in this contemporary moment. More specifically, how do these technologies, and their usages, inform our understanding of the "production" of environmental disasters, their (re)mediation, and the assignment of blame?

First, what are the media through which we are increasingly coming to remotely know this expanding place that is the disaster area? Due to the spatial and temporal scales of the spill (i.e. how much and how fast it is growing) as well as its distant location (both from land and below sea), remote sensing technologies have been of incredible value and used widely. Remote sensing is frequently thought to pertain solely to satellite images. However, as wiki informs us: remote sensing is "the use of either recording or real-time sensing device(s) that are wireless, or not in physical or intimate contact with the object". With this definition, we see that much of our individual and collective knowledge of this place, and its recent transformations, have occurred as a result of remote sensing technologies: principally satellite images and underwater video feeds.

In trying to keep this post short, I´ll simply direct the interested reader to several sites I think are thought provoking on the topic.

The first link takes you to NASA´s page devoted to the oil spill. Here you´ll find a variety of links to specific sensors, and stories derived from them (think how do we know we have environmental problems...)

The next link takes you once again to NASA, but this time o it´s AVIRIS sensor page (for images obtaned from an airplane vis a vis a sensor)


Following that you might be interested in what some of the other private and commercial satellites are up to in this time of environmental disaster. To help answer that question see the following link for imager from Terra, Digital Globe etc.

After that, on our tour of remote sensing media, I´d suggest a peek at this link to a real-time server illustration showing the growing extent of the damage showing spill.

It's not only the spatial and environmental scale of the disaster that are expanding (clearly the two are linked)as the following link shows, the political extent of spill is increasing...

Perhaps Obama gave Discover Magazine it´s headline illustrated in this last link...

What will be the short and long term implications of this spill at various scales from environmental, economic, political, and cultural perspectives. I don´t have the answers, but what I do think is that remote sensing technologies will play an important role in how we understand the nature of the problem and the potential solutions.

Geotagging photos

If a picture is worth a thousand words, as the adage goes, than what is a geotagged photo worth? Priceless. What is a "geotagged photo" you might ask....well, in borrowing shamelessly from wikipedia:

Geotagging is the process of adding geographical identification metadata to various media such as photographs, video, websites, or RSS feeds and is a form of geospatial metadata. These data usually consist of latitude and longitude coordinates, though they can also include altitude, bearing, distance, accuracy data, and place names.

Some cell phones like the iPhone and Samsung Memoir already utilize a GPS chip along with built-in cameras to allow users to automatically geotag photos. A few cameras such as the Ricoh 500SE and the Sony DSC-HX5V have built-on or built-in GPS that allow for automatic geotagging. Nikon and Canon have also come out with custom geotagging solutions. Almost any digital camera can be coupled with a GPS and post processed with photo mapping software such as GPS-Photo Link, MediaMapper, Topofusion or EveryTrail GPS Connector (for Garmin gps products) to geotag photos by matching gps coordinates with photos. Twitter, the popular social networking and microblogging service that enables its users to send and read messages known as tweets, allows its users to geotag their locations via their tweets and pictures.......


So that's geotagging in a nutshell. Now why do I consider it priceless? From the perspective of an anthropologist who strives to integrate cultural data into geographic information systems (GIS), geotagging photos is an incredible way to add texture and nuance to traditionally two-dimensional maps. Photos taken at one's field site are a form of ground truth data, or data that are collected on location (as opposed to data collected remotely, such as aerial photographs and satellite images). Ground truthed data are important as they improve the accuracy of one's analysis (regardless of whether one is a forester, anthropologist, human geographer, or urban planner).

I have become interested in geotagging photos in the context of my dissertation research on landscape change within the Brazilian Landless Workers' Movement (MST). More specifically, I think geotagging photos will have great applicability in my research for helping to tell local people's stories.


I've recently been experimenting with the Ricoh Caplio 500SE camera system (camera + integrated GPS device). To date, I've been moderately impressed with this system. Yes, it does it's job, i.e. georeferencing photos....but it is fairly complicated to figure out how to use it. What has been even more frustrating has been integrating these data into a geographic information system, which from my perspective is the entire point of these cameras. As indicated in the Wiki description, there are various programs (both online and ones that can be downloaded to your computer) that allow one to do approximate geotagging via landmarks (i.e. take a picture of your house, go into Google maps, and link that photo with your house, or get the lat/long from Google Earth and do it more accurately). However, if you want to put your photos into a geographic information system (GIS) for whatever purpose (research, urban planning, ultimate techie street-cred) than it seemed that you would need to cough up more than $300 for the GPS Photolink Pro software package. This seemed ridiculous to me, and so I did a bunch of searching online, and found a solution, which I thought I'd share with y'all: ArcPhoto

In brief, ArcPhoto is a free toolset (a script) that can be used in ArcGIS. Here is what ESRI has to say about it: The ArcPhoto Tools are a set of geoprocessing tools and ArcMap user interface enhancements to enable the quick import of digital photography into the ArcGIS framework. The tools work directly off the EXIF (Exchangeable image file format) header information that is encoded into digital imagery. This information usually includes information about the type of camera used, focal length, aperture, type of flash, etc. For professional grade cameras or images taken with ArcPad 7/8, the imagery can potentially hold GPS location information about where the photo was taken as well. At 9.2 the ArcGIS framework can directly read this type of metadata and with the help of the location information the ArcPhoto tools allow for a quick and streamlined integration of rich visual information.

Special functionality for ArcMap includes the display of thumbnail images as “map tips” and the ability to create ArcPhoto Elements – think of thumbnail images as annotation for point features. Finally within the 3D environment the images are applied to create billboards and they are displayed using the captured orientation of the digital photography.

If you download it, make sure you read through the Read Me file (which should be located here on your computer once you download it C:/Program Files/ESRI/ArcPhoto Tools/Documentation/ArcPhoto.htm).

After some brief experimenting, I found that this program provides the functionality I'm looking for, namely, taking photos (either georeferenced or not-i.e. with knowledge of lat/long points you can geotag them yourself) and integrating them in a GIS, and the ability to create shapefiles from them.

As my research progresses I'll be sure to provide updates on these technologies.